The Relative Influence of Retention, Uptake, and Translocation on the

Chapter 6

Downloaded by KANSAS STATE UNIV on November 16, 2014 | http://pubs.acs.org Publication Date (Web): October 15, 2014 | doi: 10.1021/bk-2014-1171.ch006

The Relative Influence of Retention, Uptake, and Translocation on the Bioefficacy of Glyphosate W. A. Forster,*,1 A. K. Pathan,1,3 M. O. Kimberley,2 K. D. Steele,1 and R. E. Gaskin1 Plant Protection ChemistryNZ, P.O. Box 6282, Rotorua 3043, New Zealand 2SCION, Private Bag 3020, Rotorua 3046, New Zealand 3Current address: Ministry for Primary Industries, P.O. Box 1340, Rotorua 3040, New Zealand *E-mail: [email protected].

1

Extensive experimentation in recent decades has assisted in developing a better understanding of the individual processes (spray adhesion, retention, uptake and translocation) that control the biological efficacy of applied herbicides. Various models are available, each usually limited to predicting the efficacy of an individual process. In this study, spray retention, uptake and translocation of glyphosate (as influenced by an organosilicone surfactant Silwet L-77®) on three plant species was determined. Non-linear regression models were used to describe the relative effects of retention, uptake and translocation processes on bioefficacy. All three variables were highly correlated, making it difficult to distinguish their relative influence on bioefficacy. Uptake and translocation were highly correlated (> than 0.98 for all species studied), with about 40-45% of the glyphosate taken up translocated for all three species. Overall the best model, explaining 85-99% variance, used retention and uptake together. It appeared that the bioefficacy of glyphosate towards barley and broccoli was 30-50% related to retention and 50-70% related to uptake, while glyphosate efficacy towards bean was entirely related to uptake and not directly related to retention. For practical purposes, it is reasonable to expect that experimental results from retention

© 2014 American Chemical Society In Retention, Uptake, and Translocation of Agrochemicals in Plants; Satchivi, et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 2014.

and uptake studies (or predictions of each from available models) considered together will give a good indication of the relative bioefficacy of different formulations for a given AI, across a wide range of systemic active ingredient (AI) and species, minimising the need for translocation measurements.


Introduction The challenges that are faced by agrichemical users have been increasing in complexity over recent years. On one hand regulators insist on reducing risk to the operator or environment (1), while on the other, increased efficiency is pursued by growers to maximise profits. These demands have the prerequisite that the biological efficacy of agrichemicals is at least retained, if not improved. The primary factors involved in the biological efficacy of pesticide sprays are: (1) deposition (as a proportion of the amount applied within the target area less drift), (2) retention (the overall capture by plants of spray droplets either on initial or subsequent impact, and after loss due to run-off), and in the case of systemic pesticides (3) uptake into the plant and (4) translocation to the site of biological action. Bioefficacy can be adversely affected by poor efficiency in any one step of the deposition–plant retention–uptake–translocation process. In order to optimise the biological efficacy of sprays, each of these factors need to be maximised, without having a detrimental effect on the other factors. For example, deposition to the target area can be maximised by eliminating drift using very large droplets, but this can lead to very low levels of spray retention, particularly on moderate through to difficult-to-wet targets (crops or weeds). Similarly, a formulation may maximise spray retention, but may have a detrimental effect on uptake of a particular systemic pesticide. It must be remembered that all of these factors are inextricably linked to determine the overall biological efficacy of sprays. Traditionally, expensive empirical field trials have been performed to determine the spray application parameters and/or formulations that provide the best overall biological efficacy, with no real underlying understanding of the role or extent to which the individual factors are affecting the result. It is now more common that laboratory screening methods are used to determine the retention, uptake and possibly translocation of agrichemical formulations, to make at least informed decisions on the best spray application parameters and/or formulations to test in field trials, reducing the need for large factorial experiments in the field. Nevertheless, there is a practical need to develop mathematical and computational models to help predict biological efficacy in terms of the complex interactions of spray deposition, retention (and coverage), uptake and translocation. Models for the individual factors involved in spray formulation efficacy do exist, but the usefulness of these models vary widely. Models for spray deposition from aerial application (2) are well advanced and well validated (3, 4), although their focus has been on spray drift. Equivalent models from ground application are less advanced, and the focus of ongoing studies (5, 6). While empirical models for adhesion (a component of retention) (7, 8) and retention 112 In Retention, Uptake, and Translocation of Agrochemicals in Plants; Satchivi, et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 2014.


(9, 10) aid our understanding, more universally useful process-driven models for spray adhesion (11, 12) and retention are currently being advanced (13–16). Although a range of useful models are available to help understand and predict uptake and translocation (17–31) they tend to be either too specific (in terms of active ingredient, surfactant, species), too simplistic or too complex and much more work is required before models are available that are both easy to use and universal. The key question is, if advanced process-driven models for these individual factors were available, how would they be integrated together to predict biological efficacy? The development of a universal spray formulation efficacy model (SFEM) is a challenging and resource intensive task that would require numerous experiments to cover, for example with herbicides, a diverse range of weed:herbicide formulation interactions. In addition, surfactant effects can be species specific. Individual surfactants may interact differently with herbicide formulations on different plant species (32). Add to that the diverse range of responses to environmental conditions (temperature, relative humidity, light intensity, water availability and nutrient levels (33–40), the variation in susceptibility to a particular herbicide among plant species (41), and within species variation due to plant growth stage (42–45), and the task appears daunting. These issues, the high diversity in intrinsic plant factors (e.g. dynamic surface and canopy characteristics) together with the availability of a range of formulations and application techniques demand complex experimental designs and high resource requirements. Despite the perceived complexities and known limitations, attempts are needed to generate an in-depth understanding of the relative importance of retention, uptake and translocation processes in order to efficiently direct resources towards a universal SFEM. The development of process-driven models (driven by physical processes and their associated physiological parameters), have the advantage over empirical models (derived from experimental measurements) in that they allow for extrapolation into new circumstances, which must be the ultimate goal. However they require that the underlying mechanisms that affect the physical and physiological parameters are well understood. In the current work, we have chosen to develop empirical models, aiming to increase our understanding of the relative importance of retention, uptake and translocation on biological efficacy. There are no such models in the literature, except for a small study by Pollicello et al. (46) who developed a simple equation relating retention and uptake to bioefficacy but did not consider translocation. The current study determined the bioefficacy of glyphosate, as it is effected by Silwet L-77® surfactant - a superspreader organosilicone adjuvant frequently used to increase the efficacy of glyphosate on a range of plant species (47), on three different plant species ranging in plant surface wettability and canopy characteristics. The objectives of the current study were (1) to study the effect of adjuvant and glyphosate concentration on spray retention, uptake, translocation and bioefficacy, (2) to relate spray retention, uptake and translocation of glyphosate to its biological performance on different plant species, and determine their relative influence on glyphosate bioefficacy and (3) to generate experimental data that can subsequently 113 In Retention, Uptake, and Translocation of Agrochemicals in Plants; Satchivi, et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 2014.

be used to (a) validate individual process-driven models such as retention and (b) aid in the development of an integrated spray formulation efficacy model (SFEM) for predicting the bioefficacy of glyphosate on a range of plant species.

Materials and Methods


Plant Material Plant materials used were barley (Hordeum vulgare L.), broccoli (Brassica oleracea L. var. Italica) and field bean (Vicia faba L.). These three species were chosen as model plants as they provide a good range of plant surface wettability and canopy characteristics for the spray retention, uptake, translocation and bioefficacy studies. Barley and field beans were raised from seeds in individual pots for a period of four weeks. Two week old broccoli seedlings, purchased from a local garden centre, were transplanted in individual pots and raised for a further 2-3 weeks. The three plant species were raised in individual pots, in Bloom seed raising mix (Yates NZ Ltd.), in a controlled environment facility at 20°C/15°C day/night temperature, 70% RH, in a 12h photoperiod with a photon flux density of approximately 500 μmol m-2 s-1. Chemicals For the retention and bioefficacy experiments, glyphosate (relative molecular mass = 169.1, Log P = -2.69) concentrate carrying 62% w/w glyphosate (technical grade, unformulated glyphosate-IPA salt, Nufarm, Australia) was applied in the absence or presence of Silwet L-77® (an organosilicone surfactant supplied by Momentive Performance Materials, Tarrytown, NY, USA) providing a range of dynamic surface tensions (23-73 mN m-1) in spray solution. In Experiment 1, 12 different treatments were applied to all three species: glyphosate at 6.2, 62 and 620 g ae ha-1, in the absence or presence of Silwet L-77 at 3 different concentrations (0.01, 0.025 and 0.1% w/v). In Experiment 2, eight different treatments were applied to broccoli and bean: glyphosate at 6.2, 62, 310 and 1,240 g ae ha-1, each in the presence of 0.2% (w/v) L-77, and 1,240 g ae ha-1 glyphosate in the absence or presence of L-77 (0.01, 0.025 and 0.1% w/v), while 10 different treatments were applied to barley: glyphosate at 6.2, 62, 310 and 620 g ae ha-1, each in the presence of 0.2% (w/v) L-77, 310 g ae ha-1 glyphosate in the presence of L-77 (0.01, 0.025 and 0.1% w/v), and 62, 310 and 620 g ae ha-1 glyphosate applied alone. Tartrazine dye (9 g L-1) was incorporated as a tracer in all spray treatments used for retention and bioefficacy studies. For the uptake and translocation assessments, an appropriate amount (equivalent of ca. 6.67 kBq; ca. 0.1% of total glyphosate mass in treatment solution) of radiolabelled glyphosate (phosphono-methyl-14C; ARC, Inc.; specific activity 1.924 GBq mmol-1; 99% purity) was incorporated into treatment solutions just prior to use. Equimolar concentrations of glyphosate acid (Monsanto, NZ) and isopropylamine (Sigma®, Sigma-Aldrich, Germany) were mixed together to form water-soluble glyphosate IPA salt (equivalent to that used in the retention 114 In Retention, Uptake, and Translocation of Agrochemicals in Plants; Satchivi, et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 2014.

experiments, and made up at the same concentrations), and appropriate quantities of the surfactant (to give the same concentrations as for the retention experiments) were added before incorporating into vials containing radiolabelled glyphosate. The same treatment solution concentrations (glyphosate and L-77) were used for all studies, ie. retention, uptake, translocation and bioefficacy.


Retention, Uptake, and Translocation Experiments For retention experiments, all species (15 replicate plants per treatment) were sprayed with treatment solutions at a nominal application rate of 200 L ha-1 through a hollow cone nozzle (TXVS-6 (Spraying Systems Co., Wheaton, USA); 220 kPa) mounted 0.5 m above 2/3 mean plant height of each species using a calibrated moving head track sprayer. Artificial targets (stainless steel plates x 6), mounted horizontally under the sprayer, were sprayed in all treatments to accurately confirm the spray volume delivered. After spraying, the plants were harvested at soil level and both plants and targets were washed immediately in a known volume of de-ionised water to recover retained dye. Absorbance of each solution (from plant or stainless steel plate) was measured using a Shimadzu 1240 UV/VIS spectrophotometer (tartrazine λ=427 nm) to determine total dye recovery. Individual plant surface areas were measured (using a LI-COR LI-3100 area meter; Li-cor Inc., Lincoln, Nebraska, USA) and retention was expressed as μl dye solution per cm2. This volume of spray contained a known amount of glyphosate. Percent plant retention was worked out from the amount of dye recovered per plant area devided by the known amount applied per unit area. For uptake and translocation experiments (five replicate plants per treatment), droplets (17 x 0.24 μl, ca. 770 μm diameter) were applied by a microsyringe to the central region on the adaxial (upper) surface of the youngest, fully-expanded leaf on individual plants within four hours of the start of the photoperiod. The droplet density and applied volume simulated a spray application rate of 200 L ha-1 (2 μl cm-2). The quantity of radiolabelled glyphosate applied to leaves in each treatment (5 plant replicates) was determined by dispensing 17 droplets directly into scintillation vials (5 replicates). Barley and broccoli are extremely difficult-to-wet plant targets. Hence, glyphosate treatments not containing L-77 were made in aqueous acetone (50% v/v, surface tension 30 mN m-1) so that droplets could be deposited on the leaf surface in uptake experiments. Similarly, the treatment consisting of 0.01% L-77 applied to barley was also made in aqueous acetone. The use of 50% acetone:water is considered to have no significant effect on the uptake of the active ingredient (48). After application, plants were returned to the controlled environment chambers (conditions as described above under “Plant material”) and harvested at 24 hours after treatment. The treated leaf was excised and washed with 2 x 4 ml 50% aqueous ethanol to recover unabsorbed glyphosate. Recoveries of radiolabelled glyphosate applied to plant surfaces, after droplet dry down, are greater than 96% using this method (49). Scintillant solution (13 ml ACS II; Amersham) was added to washings and radioactivity was quantified by liquid scintillation counting (Packard Tricarb 2100 TR). Foliar uptake was defined as the radioactivity not recovered from washing the treated leaves and was calculated as a percentage of the applied dose. Glyphosate mass 115 In Retention, Uptake, and Translocation of Agrochemicals in Plants; Satchivi, et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 2014.


uptake and translocation per unit plant leaf area (μg cm-2) was calculated on the basis of total glyphosate mass retention per unit plant leaf area (μg cm-2), i.e. % uptake from the uptake (or translocation) experiments multiplied by μg cm-2 retained = μg cm-2 taken up (or translocated). The washed and excised treated leaf was stored frozen in sealed plastic bags until processed in a Biological Oxidiser (Harvey OX500). Radiolabelled glyphosate within plant tissues was determined by combusting/oxidising the leaf and absorbing the 14CO2 generated in scintillant solution (Carbon-14 scintillant; RJ Harvey Corporation), which was then quantified by liquid scintillation counting. Combusting treated leaves (after surface washing) determined the amount of glyphosate which was absorbed into, but not translocated out of the treated leaf. By combining the radioactivity recovered from washings and treated leaf combustions, the amount of translocation out of the treated leaf was determined by difference. These calculations are possible because all glyphosate applied can be recovered when the tissues of whole plants are analysed ((50); uptake studied over 72 hours into Abutilon theophrasti). Bioefficacy Assessments One set of 15 plants per treatment was sprayed for bioefficacy assessments, as per the retention experiments. After spraying, plants used for bioefficacy experiments were returned to the controlled environment facility for 48 hours to allow the uptake process to occur under controlled conditions. The plants were then taken either to a glasshouse, to protect plants from frost damage and keep them in relatively warm conditions during winter months, or to an open nursery during warmer parts of the year. Bioefficacy was assessed on a 0-10 scale (0 being a healthy plant and 10 being a dead plant) by recording visible herbicide-induced damage on plants at approximately weekly intervals for up to 7 weeks. The score taken 2 weeks after spraying showed treatment effects most clearly and this score (converted to percent) was therefore chosen for analysis. The other variable used to assess bioefficacy was plant dry weight. Individual plants were excised at ground level after final scoring and placed in paper bags in a 65°C oven for 48-72 hours prior to final dry weight determinations. In addition to the sprayed plants, 15 untreated control plants per species were assessed using the same procedures in each experiment. Statistical Analysis All of the glyphosate x L-77 treatment combinations (4 x 5 factorial design) were tested for each species in the retention, uptake and translocation trials, with an additional control treatment in the bioefficacy trial. Because of the logistical limitations of handling all treatments in a single trial, the treatment combinations were split between two experiments conducted at different times for each species, using a 4 x 3 factorial design in the first experiment and a 5 x 4 factorial design in the second experiment. The two trials were essentially disconnected designs, apart from having a common control treatment for the bioefficacy assessments, and having two treatment combinations from the first trial repeated in the second 116 In Retention, Uptake, and Translocation of Agrochemicals in Plants; Satchivi, et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 2014.


experiment for barley only. Means of retention, uptake and translocation (μg cm-2) and of the bioefficacy score and final dry weight were calculated for each treatment combination. These means were then plotted against L-77 and glyphosate concentration to determine the forms of the responses. Based on these plots, appropriate nonlinear regression models were then fitted to the treatment means using the SAS (Version 9.1) NLIN procedure. Separate models were fitted for each plant species. The statistical significance of the effect of glyphosate and L-77 concentration was determined by testing relevant model parameters at the 5% level of significance. To account for differences between experiments (e.g., due to differences in plant size), the models included separate intercept parameters for each experiment fitted using dummy variables.

Models To Explain the Effect of Glyphosate and L-77 Concentration on Retention, Uptake, and Translocation The following regression model was used for modelling retention, uptake, and translocation (Y; μg cm-2) as a function of glyphosate concentration (G; %):

The model contains separate intercept parameters a1 and a2 for each experiment fitted using dummy variables. These allow for differences in leaf size or shape between experiments. The b parameter indicates the form of the relationship. For example, if b does not differ significantly from zero, it can be concluded that there is no significant relationship with glyphosate concentration. If b is close to one, it can be concluded that Y increases proportionately with glyphosate concentration, while if b is greater than 1 it can be concluded that Y increases disproportionately with glyphosate concentration. Next, the effect of L-77 concentration was incorporated into Model 1 as a multiplier (M) of the following form:

where, L is L-77 concentration (%). The coefficients in Model 2 indicate whether the L-77 enhances (if c>0) or diminishes (if c

The Relative Influence of Retention, Uptake, and Translocation on the

Recommend Documents